Gene expression signatures that define the common categories of non-Hodgkin's lymphoma (NHL) have been identified, including some unique subtypes not recognized previously. Molecular prognosticators for these NHLs have also been described. We have recently developed a DNA microarray with over 2500 features, including all known diagnostic and prognostic parameters for NHL. We propose to use this array to study 2400 new cases to validate and refine current as well as newly discovered signatures and algorithms as the study progresses. Completion of this aim will provide a robust, cost-effective system for accurate diagnostic and prognostic applications. An alternative, quantitive(Q) RT-PCR based platform that can be applied to paraffin-embedded tissue will also be developed. It will incorporate the essential information from of the microarray platform and allow the assessment of critical parameters in specimens not specifically collected for microarray analysis, hence greatly expanding the utility of the information. There is evidence that genetic data may provide additional prognostic information to the current gene expression-derived prognosticators. We propose to obtain comprehensive, high-resolution molecular genetic data on diffuse large B-cell lymphoma to examine if the combination of genetic and gene expression parameters will improve the current prognostic model. Since existing prognosticators were derived from the study of archival cases prior to the availablility of anti-CD20 therapy, they will be re-evaluated in patients treated with current therapeutic regimens. Modifications will be made to maintain the accuracy of the prognosticators. This study will not only validate previously-defined diagnostic and prognostic signatures, but will further refine the signatures and algorithms for higher accuracy and robustness. The range of application will be expanded by adapting the information to a Q-RT-PCR platform. Prognosticators will be enhanced by the incorporation of genetic data, and updated by a re-evaluation of cohorts of patients treated with current therapeutic regimens.